Getting to 99% Accuracy in Interactive Segmentation
arXiv:2003.07932
Abstract
Interactive object cutout tools are the cornerstone of the image editing workflow. Recent deep-learning based interactive segmentation algorithms have made significant progress in handling complex images and rough binary selections can typically be obtained with just a few clicks. Yet, deep learning techniques tend to plateau once this rough selection has been reached. In this work, we interpret this plateau as the inability of current algorithms to sufficiently leverage each user interaction and also as the limitations of current training/testing datasets. We propose a novel interactive architecture and a novel training scheme that are both tailored to better exploit the user workflow. We also show that significant improvements can be further gained by introducing a synthetic training dataset that is specifically designed for complex object boundaries. Comprehensive experiments support our approach, and our network achieves state of the art performance.
arXiv commentSubmitted for review to Signal Processing: Image Communication
References in corpus (6)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials
- DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs
- On the Variance of the Adaptive Learning Rate and Beyond
- Micro-Batch Training with Batch-Channel Normalization and Weight Standardization
- Interactive Video Object Segmentation in the Wild